Xref: utzoo comp.groupware:611 comp.infosystems:283
Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!usc!ucla-cs!twinsun!eggert
From: eggert@twinsun.com (Paul Eggert)
Newsgroups: comp.groupware,comp.infosystems
Subject: Re: Summary and review of ``Reading All About Computerization''
Message-ID: <1991Jun22.030814.320@twinsun.com>
Date: 22 Jun 91 03:08:14 GMT
References: <9106191945.aa10061@ics.uci.edu>
Sender: usenet@twinsun.com
Organization: Twin Sun, Inc
Lines: 31
Nntp-Posting-Host: ata

>I am very curious to know which "half dozen other"
>classification schemes you actually find much more useful,
>and specificlly in what ways and why.

I didn't say ``much more useful'', I just asked why the paper's
classification was any better.  Here are six schemes off the top of my
head, together with brief examples of their use.

expert vs novice _audience_
  One can usually ignore expert material if one is a novice, and vice versa.

expert vs novice _writer_
  One can usually ignore writers who know little about the subject.

policy vs mechanism vs verification
  Policy is ``What should the system do?''; mechanism is ``What
  technical means should it use?''; verification is ``How do we know
  the mechanism implements the policy?''.  These issues are all
  important at different stages of building a collaborative system;
  distinguishing them clarifies design and avoids pointless arguments.

short-term vs long-term view
  One can usually ignore old literature that talks about short-term issues.

technology-driven vs market-driven
  One can usually ignore market research if one is trying to derive
  a technological solution to a known problem; and vice versa.

Who is the sponsor?
  One should not be surprised when an author from organization X
  evaluates X's system positively.